What exactly is "Pull" type inventory planning?

It has been years since I read the notion of a 'Pull' type planning where a manufacturer do not  build any inventory and wait for a customer order to start its process. However it is also true that in several instances this is not possible, because customer is not going to wait through your lead time for the product to show up. What does this mean? Does it mean that the pull technique is ONLY applicable for a "make to order" environment! If this is the case, than even Toyota do not seem to be following it. One can visit any Toyota dealer in US to see several cars readily available for the customers. Does it mean that they do not practice what they preach? I had this question for years before I found the answer recently and here is how I see it now. 

It is important to bring a nice historical perspective to the Toyota story. As the story goes, manager's from Toyota came to US to learn about Auto industry. What they noticed in Ford did not exactly impressed them! However they were fascinated by a Wrigley grocery store where they would stock a few items in the isle based on historical consumption (aka actual demand) and replenish them whenever those isles are empty. The inventory were often minimal and based on exact demand. 

The management of inventory planning process at Wrigley happened to be the key motivator for Toyota. They created the pull based system by storing just a minimum inventory at each desired location. The amount was based not on any forecast, but on actual consumption of those items. The replenishment order was initiated not based on any pre-determined schedule, but on actual drop in inventory below the set level. This inventory was aptly called the "Supermarket" in the language of lean. As the inventory was pulled by customers, replenishment orders (Work order or Purchase order) were placed on the downstream suppliers. And this is how a pull based planning can be made functional even in a "make to stock" environment. 

So pulling does not mean a "make to stock" or a "make to order" planning; it means the ability to set the inventory based on actual demand (measured by historical consumption) and not on any forecast. It also means that the replenishment (production or purchase) is based not on any schedule, but on actual need for the inventory to support the demand. Forecast is no-longer used to guide the production process. 

But can it be true when future is very different from past!! If you have a very strong reason to believe that forecast is going to be true, than use it (instead) to set the inventory. This can happen when a customer indicates about a large order in advance. So inventory is usually set based on actual demand; however it can also be set at a higher/lower level when the planners have strong reason to believe that the forecast is going to be true!

Additionally, in order to take care of the uncertainty we add a "safety stock" component to the inventory. This is also calculated using the historical actual demand data.  

Comments are welcome!
Print this Article

Incentive mis-match in retail, what the OEM's can do?

What is the Problem:

As margins of the big-box electronic retailers like Circuit City & Best Buy started shrinking, they tried to enter the services business through initiatives like "Geek Squad". This per se is not a unique idea; OEM's like SUN have long tried use this services business as their prime driver for profit. These services are often sold along with products. However retailers getting into this services business is a relatively recent phenomenon. When a customer is buying a NEW Canon digital camera, he would probably not buy a 2~3 years of maintenance contract unless he/she is convinced that there is a reasonable likelihood that the camera is going to breakdown in the next couple of years. Hence the services are often sold to the customers with the presumption that the product may not be good enough to remain trouble free for a couple of years. This also creates a platform for the large retailers to prepare and enter the market with their own more profitable "private label" when they feel that the consumer electronics has become a commodity. But this is not a very good promotion for Canon Digital Camera or a Sony Plasma TV; products which more or less have an excellent trouble free life at least for the first few years. Retailers instead of promoting the quality and longevity of the products are inadvertently indicating the lack of it. It is a good deal for customers, because if et all their equipment really fails, they are better off taking it to the nearest retailer instead of sending it to some remote repair facilities of the OEM and waiting for a longer uncertain period. So the services provided by the retailers are intuitively attractive to the customers if they have to have a service contract. Hence from a customer perspective, they would rather buy it from a retailer in stead of the manufacturer itself.

All these are pointing towards a win-win situation for both retailers and the customers at the expense of OEM's whose product life and quality to certain extent are being openly questioned. Now what should the OEM do to avoid this trap? It is important to remember here that due to the size and reach, these large retailers have a strong bargaining power and at the same time, electronics major are noticing a steady decline in their margin with a continuing trend in commoditization of products.

I have noticed this problem several years ago and do not have any idea on how this can be solved from an OEM perspective. Any comments on this is most welcome!

Comments are welcome!
Print this Article